NONINVASIVE ANALYSIS OF CARTILAGE VOLUME AND THICKNESS IN THE HUMAN ELBOW JOINT WITH MRT

The aim of the study was to non-invasively analyse the cartilage volum e and thickness in the human elbow joint with magnetic resonance imagi ng. 12 fresh frozen specimens (ages 20 to 69 yrs.) were investigated u sing a 1.5 T magnet, and a water-excitation FLASH-3D sequence at a res olution of 1 x0.25x0.25 mm(3). After linear interpolation to 0.125x0.1 25 mm(2) in the image plane, the cartilages were segmented interactive ly with a Snake algorithm. Following three-dimensional reconstruction, the cartilage volumes were determined, and the mean and maximal carti lage thickness computed by Euclidean distance transformation. The tota l cartilage volume of the human elbow joint amounted to between 3.90 a nd 7.17 mi (mean 5.5 ml +/- 20%). The humerus occupied 49 to 60%, the radius 15 to 27% and the ulna 20 to 29% of the total volume. The mean cartilage thickness ranged from an average of 0.9 (proximal part of th e ulna) to 1.4 mm (capitulum humeri), and the maximal thickness from 2 .3 mm (proximal part of the ulna) to 2.9 mm (distal part of the ulna). The ulnar cartilage showed a more inhomogeneous distribution than tha t of the humerus and radius. The interindividual variability of the ca rtilage thickness was less than that of the volume. There was no signi ficant relationship of the volume with age (r = 0.11) or body weight ( r=0.51). However, based on the joint size (r=0.71-medio-lateral extens ion of the articular surfaces) about 50% of the variability of the tot al cartilage volume could be predicted. The technique presented is sui table for designing computer models to investigate the load transmissi on and functional adaptation of diarthrodial joints, and for diagnosin g and monitoring joint disease.